⚠️ kubeflow/examples Repository is not Maintained

This repository is no longer actively maintained, and some examples may be outdated or non-functional. For the working examples, please refer to the GitHub repositories of the individual Kubeflow components.

If you are interested in contributing to the kubeflow/examples repository, we encourage you to join the Kubeflow community calls and share your interest.

Help wanted blog post

Blog post: HELP WANTED: Repackaging Kaggle Getting Started into Kubeflow Examples

higlights:

• We'd like to help bolster the kubeflow/examples repo
• Help people get involved in open source/kubeflow project/community
• Give people an opportunity to make a little side hustle income

Kubeflow Examples

A repository to share extended Kubeflow examples and tutorials to demonstrate machine learning concepts, data science workflows, and Kubeflow deployments. The examples illustrate the happy path, acting as a starting point for new users and a reference guide for experienced users.

This repository is home to the following types of examples and demos:

• End-to-end
• Component-focused
• Demos

End-to-end

Named Entity Recognition

Author: Sascha Heyer

This example covers the following concepts:

1. Build reusable pipeline components
2. Run Kubeflow Pipelines with Jupyter notebooks
3. Train a Named Entity Recognition model on a Kubernetes cluster
4. Deploy a Keras model to AI Platform
5. Use Kubeflow metrics
6. Use Kubeflow visualizations

GitHub issue summarization

Author: Hamel Husain

This example covers the following concepts:

1. Natural Language Processing (NLP) with Keras and Tensorflow
2. Connecting to Jupyterhub
3. Shared persistent storage
4. Training a Tensorflow model
   1. CPU
   2. GPU
5. Serving with Seldon Core
6. Flask front-end

Pachyderm Example - GitHub issue summarization

Author: Nick Harvey & Daniel Whitenack

This example covers the following concepts:

1. A production pipeline for pre-processing, training, and model export
2. CI/CD for model binaries, building and deploying a docker image for serving in Seldon
3. Full tracking of what data produced which model, and what model is being used for inference
4. Automatic updates of models based on changes to training data or code
5. Training with single node Tensorflow and distributed TF-jobs

Pytorch MNIST

Author: David Sabater

This example covers the following concepts:

1. Distributed Data Parallel (DDP) training with Pytorch on CPU and GPU
2. Shared persistent storage
3. Training a Pytorch model
   1. CPU
   2. GPU
4. Serving with Seldon Core
5. Flask front-end

MNIST

Author: Elson Rodriguez

This example covers the following concepts:

1. Image recognition of handwritten digits
2. S3 storage
3. Training automation with Argo
4. Monitoring with Argo UI and Tensorboard
5. Serving with Tensorflow

Distributed Object Detection

Author: Daniel Castellanos

This example covers the following concepts:

1. Gathering and preparing the data for model training using K8s jobs
2. Using Kubeflow tf-job and tf-operator to launch a distributed object training job
3. Serving the model through Kubeflow's tf-serving

Financial Time Series

Author: Sven Degroote

This example covers the following concepts:

1. Deploying Kubeflow to a GKE cluster
2. Exploration via JupyterHub (prospect data, preprocess data, develop ML model)
3. Training several tensorflow models at scale with TF-jobs
4. Deploy and serve with TF-serving
5. Iterate training and serving
6. Training on GPU
7. Using Kubeflow Pipelines to automate ML workflow

Pipelines

Simple notebook pipeline

Author: Zane Durante

This example covers the following concepts:

1. How to create pipeline components from python functions in jupyter notebook
2. How to compile and run a pipeline from jupyter notebook

MNIST Pipelines

Author: Dan Sanche and Jin Chi He

This example covers the following concepts:

1. Run MNIST Pipelines sample on a Google Cloud Platform (GCP).
2. Run MNIST Pipelines sample for on premises cluster.

Component-focused

XGBoost - Ames housing price prediction

Author: Puneith Kaul

This example covers the following concepts:

1. Training an XGBoost model
2. Shared persistent storage
3. GCS and GKE
4. Serving with Seldon Core

Demos

Demos are for showing Kubeflow or one of its components publicly, with the intent of highlighting product vision, not necessarily teaching. In contrast, the goal of the examples is to provide a self-guided walkthrough of Kubeflow or one of its components, for the purpose of teaching you how to install and use the product.

In an example, all commands should be embedded in the process and explained. In a demo, most details should be done behind the scenes, to optimize for on-stage rhythm and limited timing.

You can find the demos in the /demos directory.

Third-party hosted

Source	Example	Description

Get Involved

• Slack
• Twitter
• Mailing List

In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, education, socio-economic status, nationality, personal appearance, race, religion, or sexual identity and orientation.

The Kubeflow community is guided by our Code of Conduct, which we encourage everybody to read before participating.